The Harris Poll^® #12, February 9, 2005

Political Labels: Majorities of U.S. Adults Have a Sense of What Conservative, Liberal, Right Wing or Left Wing Means, But Many Do Not

Journalists, political activists, and many others routinely use political labels to describe politicians and policies as conservative or liberal, right-wing or left-wing. A new Harris Poll measures what a cross-section of U.S. adults understand by these and other labels.

Most people, it appears, understand these labels in pretty much the same way political pundits do. Large majorities believe that conservatives favor moral values, cutting taxes, and oppose same-sex marriage, gay rights, and abortion rights. Majorities believe liberals favor abortion rights, gay rights, same-sex marriages, and affirmative action. But substantial numbers of people don’t know where conservatives and liberals stand on those and other issues. And some people seem to completely misunderstand these labels.

These are the results of a new Harris Poll of 2,209 U.S. adults surveyed online by Harris Interactive^® between January 11 and 16, 2005.

While most people gave the expected answers, substantial minorities think that conservatives oppose cutting taxes (19%) or are not sure (11%) whether they favor or oppose cutting taxes. In other words, 30 percent of all adults do not give the expected response that conservatives favor cutting taxes. Similarly:

• 50 percent believe that conservatives support gun control or are not sure.
• 46 percent think that conservatives support affirmative action or are not sure.
• 23 percent think that conservatives support abortion rights or are not sure.
• 19 percent think that conservatives support gay rights or are not sure.
• 15 percent believe conservatives support same-sex marriage or are not sure.

While large majorities believe liberals favor abortion rights, gay rights, and same-sex marriage, substantiated minorities give more surprising responses:

• Fully 39 percent believe liberals favor cutting taxes, and 17 percent are not sure.
• Fully 37 percent believe that liberals either oppose gun control (24%) or are not sure (13%).
• Significant but smaller numbers do not believe or are not sure if liberals support gun control (37%), affirmative action (26%), same-sex marriage (22%), gay rights (17%), or abortion rights (16%).

As one would expect, many people think that right-wingers support and oppose policies in ways which are similar to the positions of conservatives. But the label right wing is less clear to many people than conservative. More people don’t know what the phrase means with between 20 and 27 percent saying they are not sure whether a right-winger would support or oppose each of the seven policy positions.

The majorities who give the expected answers (e.g. the 59 percent who say that right-wingers support cutting taxes) are about 10 points lower than those who give the expected answers when describing conservatives.

Likewise, the label left-winger is broadly perceived to be similar to liberal except that more people are not sure what it means.

Unsurprisingly, perceptions of moderates and independents fall between those of conservatives and liberals and between right and left-wingers. But there are some interesting, possibly surprising findings.

Moderates for example are seen by many people to resemble conservatives in supporting moral values and tax cuts but to resemble liberals in supporting abortion rights, gun control and gay rights. Smaller numbers, but still pluralities, believe that independents favor tax cuts, abortion rights, affirmative action, gay rights, gun control and same sex marriage.

Overall, therefore, people are more likely to see moderates and independents as resembling liberals than resembling conservatives.

Harris Interactive is indebted to the distinguished social scientist, Leo Bogart, who suggested that we measure what political labels mean to people. After reviewing these findings, Bogart, commented that, "The confusion over what the widely used political labels actually mean reflects their common use as epithets rather than as true descriptions of people's beliefs. Those beliefs are often inconsistent and self-contradictory, because Americans don't adhere to a party line or doctrine that provides ready answers to all questions. We may be "conservative" on some issues and "liberal" on others, but it's the specifics that

matter, not the labels."

"Thinking now of the political climate in the United States, do you think of conservatives as people who support or oppose…?"

Base: All adults

"Thinking now of the political climate in the United States, do you think of liberals as people who support or oppose…?"

Base: All adults

Note: Percentages may not add up exactly to 100% due to rounding.

"Thinking now of the political climate in the United States, do you think of right-wingers as people who support or oppose…?"

Base: All adults

Note: Percentages may not add up exactly to 100% due to rounding.

"Thinking now of the political climate in the United States, do you think of left-wingers as people who support or oppose…?"

Base: All adults

Note: Percentages may not add up exactly to 100% due to rounding.

"Thinking now of the political climate in the United States, do you think of moderates as people who support or oppose…?"

Base: All adults

Note: Percentages may not add up exactly to 100% due to rounding.

"Thinking now of the political climate in the United States, do you think of independents as people who support or oppose…?"

Base: All adults

Note: Percentages may not add up exactly to 100% due to rounding.

The Harris Poll^® was conducted online within the United States between January 11 and 16, 2005 among a nationwide cross section of 2,209 adults aged 18 and over. Figures for age, sex, race, education, region and household income were weighted where necessary to bring them into line with their actual proportions in the population. Propensity score weighting was also used to adjust for respondents’ propensity to be online.

In theory, with probability samples of this size, one could say with 95 percent certainty that the results have a sampling error of plus or minus 2 percentage points. Unfortunately, there are several other possible sources of error in all polls or surveys that are probably more serious than theoretical calculations of sampling error. They include refusals to be interviewed (non-response), question wording and question order, and weighting. It is impossible to quantify the errors that may result from these factors. This online sample is not a probability sample.

J23069

Q810, Q820